Individual differences in DNA methylation are associated with cardiovascular disease (CVD) risk factors. Our group focuses on a class of genomic regions called `metastable epialleles' (MEs). MEs are loci where methylation is `set' during very early embryogenesis in response to non-genetic factors such as maternal nutrition. As such, the degree of hypo- or hyper- methylation at MEs is systemic and so detectable in all tissues, and it is stable across the lifespan. Our group has been at the forefront of inbred mice studies which revealed permanent consequences of individual variation in methylation at MEs on phenotypes such as obesity. Now we need to determine the phenotypic effects of methylation at MEs in humans. Monozygotic (MZ; identical) twins are commonly compared to dizygotic (DZ; fraternal) twins in studies seeking to separate genetic from non-genetic influences on methylation. However, in contrast to inbred mice, most MZ twins share a placenta in utero. This leads to intermingled circulation, and the cross-pollination of hematopoietic stem cells. We hypothesize that this leads to higher MZ concordance in peripheral blood cells (PBCs) and saliva (which contains PBCs) than in other somatic tissues. The overall goal of the proposed investigation is provide data which inform the design of twin studies that examine the effects of methylation at MEs on CVD risk. As infancy is a period close to embryogenesis and offers one of the earliest periods for risk stratification and intervention, we will leverage the existing structure of The Baylor Infant Twin Study (BITS) and recruit infants at ages four months or younger. We will collect PBCs, urine, fingernail clippings and saliva from buccal swabs on 80 MZ infant twin pairs, and collect hair follicles on a subsample of 23 infants. We will conduct pyrosequencing to determine the % methylation at 10 established ME loci, in genomic regions associated with CVD risk factors. The first aim of this study will be to validate nail clippings as an acceptable DNA source for the study of DNA methylation in humans. We anticipate that methylation at 10 ME loci will be comparable (r?.71) between nail clippings and hair follicles. The second aim will compare methylation discordance between MZ twins who shared a placenta in utero (monochorionic), and MZ twins with separate placentae (dichorionic) across 10 ME loci. We expect that monochorionic twins will show a higher concordance than dichorionic twins in tissues which contain hematopoietic stem cells (PBCs, saliva), but not in those which do not (fingernails and urine). This study will provide data that informs subsequent ME study design. Such studies will seek to disentangle genetic and environmental effects on methylation established during embryogenesis, and reveal how this influences CVD risk. Ultimately, as methylation at MEs is established in response to periconceptional factors such as maternal nutrition, this R21 proposal will form the first step in program of research which aims to provide a new intervention target for reducing CVD risk.